Open data

Definition and delimitation

Open data refers to data that is freely accessible, usable, modifiable and shareable by anyone for any purpose – restricted at most by measures that preserve the origin and openness of the knowledge ^[1]. Open data usually refers to public sector information (PSI), i.e. all “information that is generated, created, collected, processed, stored, preserved, disseminated or financed by or for a government institution or public body” ^[2]and in this context is also referred to as Open Government Data. According to a broad understanding of the term, Open Data can also refer to access to data owned by private actors and companies.

History

In the European Union, the first binding legislation on Open Data was adopted in 2003: Directive 2003/98/EC on the re-use of public sector information (PSI Directive) aimed to harmonise the publication of public sector information between European member states and thereby reduce barriers to the re-use of this information ^[1]. In particular, the Directive prohibited exclusive contracts regarding the provision of data by public sector bodies to third parties (subject to specific exceptions), required general non-discriminatory access to public sector information for all potential re-users and promoted its publication on online portals to facilitate access and discovery by third parties. In 2013, the directive was revised and supplemented. Among other changes, the revised version of the directive emphasised in particular the need to provide information in a machine-readable format. In addition, a general upper limit for access fees to Open Data was set at the level of marginal costs, which could, however, be exceeded in exceptional cases.

As part of the initiative to “build a common European data space”, the PSI Directive was revised again in 2019 and adopted as Directive (EU) 2019/1024 on open data and the re-use of public sector information (Open Data Directive). The recast directive entered into force on 16 July 2019. The implementation period for member states of two years ended on 17 July 2021. The directives oblige member states to make all existing documents (in particular also electronically stored data) re-usable, unless this is prevented by national regulations or exceptions regulated in the directive ^[2]. Compared to the PSI Directive, the Open Data Directive further restricts the use cases in which access fees above marginal cost may be charged, increases the transparency requirements for agreements between public and private institutions on data use and emphasises the need to make dynamic data accessible in real time via application programming interfaces (APIs). Special attention has also been paid to small and medium-sized enterprises (SMEs), which should benefit from the use of Open Data as easily as possible. Finally, the directive stipulates the definition of high-quality data sets “whose re-use is associated with important benefits for society, the environment and the economy” (Art. 2 (10) of the Open Data Directive). Due to the special value potential of these data, they are to be made available free of charge, in a machine-readable format, via APIs and, where appropriate, as a mass download (Art. 13). While the Open Data Directive defines six thematic categories of high value datasets, the final list of datasets will be determined by the European Commission in additional implementing acts. On 10 February 2021, the German Federal Government adopted the draft law on the use of public sector data, which is intended to implement the European Directive nationally.

Application and examples

Open Data is provided and used in many different forms. In particular, the provision and use of public data is intended to enable data-driven innovation, strengthen the competitiveness of small and medium-sized enterprises and promote non-profit goals such as transparency and accountability of government activities. Open data and the information it contains can directly concern government activities, e.g. in the form of legal texts or information on public institutions. However, government agencies also collect information on the private sector, such as demographic population information or business registers. Finally, the public sector collects a lot of environmental information, often in high quality and high detail.

In commercial use, Open Data can be used by companies, on the one hand, directly as an “end product” to create added business value. This means that the information contained in the data is presented in a meaningful context or visualised in a way that creates added value for customers. To get the necessary information, companies typically collect data from their users, scrape web content or gather data from offline sources. Open Data provides an additional source for such services. For example, the HD Scores application uses data from the US Department of Health and Human Services to inform the public about the quality of restaurants ^[4]. Another prominent case of representative use of information based on OGD is the display of satellite data from NASA in the map displays of Google Maps ^[5]. In addition, Google uses various other OGD sets such as weather and regional data to build and enrich the Google Maps service ^[6].

On the other hand, Open Data can be used by companies as an “intermediate product” when the data serves as input for data analytics applications and data-driven decision support systems. In this way, Open Data can increase productivity and efficiency of an organisation. Specifically, Open Data can be used in the context of business intelligence to conduct market analyses, create industry benchmarks or improve risk management models ^[7]. In addition, companies also use Open Data to optimise internal business processes or personnel planning with a view to demand trends ^[8]. In the context of digital services, data analytics applications play an important role in deriving new knowledge and making predictions, as well as improving service quality through targeted personalisation and appropriate delivery of content and advertising. Open Data can thus contribute to the development of completely new product or service offerings and in this way promote digital innovation ^[7]. In addition, some companies are using Open Data and the demographic information it contains to better tailor individual service offerings and targeted advertising to user interests ^[9].

Criticism and problems

With regard to Open Data, there is regular debate about the extent to which public sector information should be made freely accessible and under what conditions this should happen. Advocates of open data access argue that the creation or collection of the data was financed with public money and that it should therefore be accessible to as broad a public as possible. Furthermore, it is pointed out that data, as a non-rivalrous good, is not used up during use and therefore the greatest possible socio-economic value can be achieved through the broadest and most diverse use of open data. It is also derived from the idea of Open Government that government activities should be as transparent and comprehensible as possible, which in turn requires the provision of Open Data.

The open provision of data can conflict with the protection of personal data or the protection of business secrets if Open Data contains information about the private sector. This is particularly the case if data is collected in particularly fine-grained units and an (indirect) identification of persons or organisations is possible. Through the state collection and provision of Open Data, the individual’s control over the publication of certain data sets may be limited. In this context, indirect economic consequences from the use of data can lead to the disadvantage of individual persons, even if there is no direct personal reference in the data.

With regard to the provision of data free of charge and without discrimination, it is argued on the one hand that the greatest possible degree of innovation can be generated in this way. On the other hand, critics point out that private-sector data-driven innovations could be robbed of their economic sales opportunities by Open Data if a state-subsidised provision of data makes commercial offers uneconomical (crowding out). In addition, in some applications, economic data use is only possible if data is exclusively accessible or others can be excluded from using it. With regard to the provision of open data, it is also criticised that it is sometimes only available in poor quality (in particular lack of completeness or lack of topicality), data sets are difficult to find or are fragmented across different institutions, and use is often only possible to a limited extent due to a lack of metadata. Access fees above marginal costs could create incentives for the provision of higher data quality at this point. Finally, it is questionable whether, due to economies of scale and scope in data use, it is not in particular already dominant and data-powerful companies that benefit from Open Data. This would run counter to the goal of promoting small and medium-sized enterprises and could further increase market concentration in digital markets.

---